Occupational asthma (OA) is one of the most prevalent occupational respiratory diseases. In spite of efficient programs allowing the early removal from exposure of the workers with OA, seventy percent of these workers remain with asthma years after removal from exposure. There is no current marker allowing the identification of the workers who may be at risk of experiencing further deterioration of their asthma after removal from exposure. A recent study suggested that the occurrence of an asthmatic reaction without an eosinophilic inflammation after exposure to the offending agent may be an indicator of a poor asthma prognosis. The broad aim of this study is to determine whether or not the lack of airway eosinophilic inflammation following the exposure to their offending occupational agents is a marker of poor asthma prognosis in workers with OA. The specific objectives are : 1. To compare the clinical, functional and inflammatory characteristics of workers with (Eos+) and without sputum eosinophilia (Eos-) after specific inhalation challenges (SIC) at the time of the diagnosis;2. To assess whether or not the non-eosinophilic phenotype shows an increase in sputum neutrophils (>60%). 3. To compare the clinical, functional and inflammatory characteristics of workers Eos+ and Eos- 5 years or more after the diagnosis. 4. To compare some markers of airway remodeling in the Eos+ and Eos- groups;5. To assess the medical resource use 5 years before and after diagnosis in Eos+ and Eos- groups. Methods: We will conduct a cross-sectional cohort study of subjects diagnosed with OA between 2000 and 2005. We will identify the charts of all subjects who had a diagnosis of OA between 2000 and 2005 in our centre and who had sputum differential cell counts performed during the investigation. These subjects will be invited to attend a follow-up visit. Spirometry, methacholine challenge, sputum induction and measure of exhaled nitric oxide (eNO) will be performed. We will also obtain the data regarding their use of medical resources (visit to the physicians, visits to the emergency and hospitalizations) 5 years before the diagnosis of OA and until the follow-up visit. The primary outcome will be the change in non-specific airway responsiveness between the diagnosis and the follow-up visit in subjects with and without eosinophilic inflammation. A sample size of 61 subjects will allow having a power of 80% with an alpha error of 5% to obtain an anticipated effect size (r2) of 0.25 in the linear regression with 6 predictors. Identifying a group of subjects at risk to have a poorer outcome may allow us to improve the management of those subjects by modifying their treatment and implementing a closer follow-up. PUBLIC HEALTH RELEVANCE: This project is highly relevant for the workers as well as for the agencies responsible for the management of the workers with occupational asthma. Indeed, if we identify a specific type of airway inflammation as being associated with a poor prognosis of the disease, this will allow an early identification of the workers at risk of a further deterioration of their asthma. This will allow a closer follow-up of these workers along with an intensification of their asthma treatment after diagnosis.